The recent announcement of HAU, a journal, "which aims to situate ethnography as the prime heuristic of anthropology, and return it to the forefront of conceptual developments in the discipline" has led me to reflect on how ethnography became so central to the practice of, in particular, social and cultural anthropology. My starting point is a framework proposed by world system theorist Immanuel Wallerstein to explain how the current constellation of social science disciplines emerged. I am still looking for the proper citation and, if anyone knows it, would be grateful to have it provided in a comment. Reader be warned: I am working from memory. Here goes.
According to Wallerstein the constellation of social science disciplines that emerged in the 20th century reflected a tripartite division of the world into (1) the advanced West, (2) the traditional East, and (3) the primitive Rest. Economics, political science, sociology and psychology divided up the study of the West along institutional lines: the Economists got the market, the political scientists the state, the sociologists civil society, and the psychologists the individuals who participate in all three. Area studies got the traditional civilizations of South Asia, the Far East, North Africa and the Middle East. Anthropology got the the Rest, which, conceived as being neither advanced nor having a rich history were conceived as primitive. It hardly needs saying that these divisions reflected a classic unilineal image of human evolution, from preliterate, primitive, little communities to traditional states/empiries, culminating in the modernity associated primarily with Europe and North America.
My purpose here is not to critique this framework but, instead, to examine its consequences for research methods and the data they analyze. I observe that only in the "advanced West" could social scientists find the statistical information collected by modern states and the social and other apparatus required for large-sample statistical surveys. Not surprisingly, it was in the West that quantitative methods became central to social science research. The availability of quantitative data was far more limited in the "traditional East." There were, however, vast libraries of chronicles, histories, and scriptures to study. As a result, historical (complemented by archeological or art historical) research became the preferred modus operandi of area studies. What, then, of the "primitive Rest"? Here neither quantitative nor text-based historical research were feasible. The only and, thus, the best available data could only be assembled by direct, face-to-face, hands-on, personal experience, a.k.a., ethnography. Thus it was, my tale concludes, that ethnography became central to social and cultural anthropology.
But that was the 20th century. What of the 21st? Now most of the globe is governed by states that generate statistical data, and the anthropologist whose field is in any of the places that once comprised the "traditional East" is, ipso facto, no longer working in a space where synchronic research was justified by the absence of historical records. My conclusion is not that the face-to-face, hands-on, personal experience that ethnography brings to the table is no longer relevant. It is, however, that it can no longer stand alone, even in places that were once seen as belonging to the "primitive Rest."
Thus, for example, I once wrote, "Malinowski's notebook made him, the anthropologist, the authority on which Trobriand spells were accurate [a claim the Malinowski himself makes; for only he was in a position to evaluate spells using the techniques of critical, textual analysis]. In 1972, Trobrianders had notebooks, too. On the night of the game Weiner (1976) describes, the winner [he who knew the most spells] was Benemiga whose "magic was written in a thick notebook that his son read to him for part of the time he chanted."
We anthropologists can no longer, therefore, present our ethnographies and say, "Those are the facts." Those are only the facts that our personal encounters have provided, and to be persuasive they must be set alongside other, statistical or historical, facts, with which they are consistent. Today's ethnographer must be comfortable analyzing relevant statistics and familiar with the history that surrounds the moment and place in which the fieldwork took place. If we wish to speak to larger issues in bigger places, we have a lot more work to do.